Erythroid differentiation is augmented in Reelin-deficient K562 cells and homozygous reeler mice

FEBS Letters 588 (2014) 58–64

journal homepage: www.FEBSLetters.org

Erythroid differentiation is augmented in Reelin-deficient K562 cells and homozygous reeler mice Hui-Chun Chu a,1, Hsing-Ying Lee a,1, Yen-Shu Huang b, Wei-Lien Tseng a Ching-Ju Yen b, Ju-Chien Cheng c,⇑, Ching-Ping Tseng a,b,d,⇑ a

Graduate Institute of Biomedical Science, College of Medicine, Chang Gung University, Kwei-Shan, Taoyuan 333, Taiwan, ROC Department of Medical Biotechnology and Laboratory Science, College of Medicine, Chang Gung University, Kwei-Shan, Taoyuan 333, Taiwan, ROC Department of Medical Laboratory Sciences and Biotechnology, China Medical University, Taichung 404, Taiwan, ROC d Molecular Medicine Research Center, Chang Gung University, Kwei-Shan, Taoyuan 333, Taiwan, ROC b c

a r t i c l e

i n f o

Article history: Received 20 December 2012 Revised 8 October 2013 Accepted 4 November 2013 Available online 12 November 2013 Edited by Ned Mantei Keywords: Erythroid differentiation K562 cell Sodium butyrate Reeler mice Reelin

a b s t r a c t Reelin is an extracellular glycoprotein that is highly conserved in mammals. In addition to its expression in the nervous system, Reelin is present in erythroid cells but its function there is unknown. We report in this study that Reelin is up-regulated during erythroid differentiation of human erythroleukemic K562 cells and is expressed in the erythroid progenitors of murine bone marrow. Reelin deficiency promotes erythroid differentiation of K562 cells and augments erythroid production in murine bone marrow. In accordance with these findings, Reelin deficiency attenuates AKT phosphorylation of the Ter119+CD71+ erythroid progenitors and alters the cell number and frequency of the progenitors at different erythroid differentiation stages. A regulatory role of Reelin in erythroid differentiation is thus defined. Ó 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction Reelin is a large extracellular glycoprotein that is mainly expressed in the nervous system. Reelin can also be detected in several other organs and tissues, including the dental pulp, liver and blood [1,2]. In the central nervous system Reelin binds to the apolipoprotein E receptor 2 and very low density lipoprotein

Abbreviations: EPO, erythropoietin; FBS, fetal bovine serum; Hct, hematocrit; Hb, hemoglobin; HRP, horseradish peroxidase; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; MCV, mean corpuscular volume; NaB, sodium butyrate; NFAT, nuclear factor of activated T cells; PAGE, polyacrylamide gel electrophoresis; PBS, phosphate-buffered saline; PVDF, polyvinylidene fluoride; RDW, red blood cell distribution width; RPMI, Roswell Park Memorial Institute; SDS, sodium dodecyl sulfate; TMB, 3,30 ,5,50 tetramethylbenzidine. ⇑ Corresponding authors. Address: Department of Medical Biotechnology and Laboratory Science, Chang Gung University, 259 Wen-Hwa 1st Road, Kwei-Shen, Taoyuan 333, Taiwan, ROC. Fax: +886 3 2118355 (C.-P. Tseng). Department of Medical Laboratory Science and Biotechnology, China Medical University, No. 91, Hsueh-Shih Road, Taichung 404, Taiwan, ROC. Fax: +886 4 22022073 (J.-C. Cheng). E-mail addresses: [email protected] (J.-C. Cheng), [email protected] (C.-P. Tseng). 1 These authors contributed equally to this work and are considered as co-first authors.

receptor, and initiates cellular signals to regulate radial neuronal morphology and cortical neuronal migration [3,4]. A number of studies indicate that Reelin is expressed in erythroid cells and haematopoietic tissues. Reelin protein is present in human erythroleukemia K562 and HEL cells [5]. Reelin transcript is induced when K562 cells undergo erythroid differentiation concomitant with mitochondria-dependent apoptosis [6]. In the developing mouse, Reelin can be detected in the somites, yolk sac, and foregut at E8.5 [7]. Reelin is one of the erythroid-enriched transcripts in primitive erythroid precursors of E9.5 yolk sac [8], and is a target gene of the transcription factor Kruppel-like factor 2, which is required for erythroid development [9]. Nevertheless, it is not clear whether Reelin has any functional role in erythroid differentiation. In this study, both K562 cells [10] and Reelin-deficient reeler mice [11] were used as models to address the role of Reelin in erythroid differentiation. Reelin was up-regulated during sodium butyrate (NaB)-induced erythroid differentiation of K562 cells. The cells with Reelin deficiency display augmentation of erythroid differentiation. Consistent with these findings, reeler mice display an increase in RBC count, hemoglobin (Hb) content, hematocrit (Hct) and mean corpuscular hemoglobin concentration (MCHC) in the peripheral blood, and augmented RBC production in bone marrow. This study thus identifies a negative regulatory role of Reelin in erythroid differentiation.

0014-5793/$36.00 Ó 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.febslet.2013.11.002

H.-C. Chu et al. / FEBS Letters 588 (2014) 58–64

59

2. Materials and methods

2.5. Real-time reverse transcription PCR

2.1. Materials

Real-time RT-PCR was performed as described in the Supplementary methods. Relative gene expression with b-actin or glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as the normalization control was determined by the 2DDCt method where Ct = threshold cycle [17]. The primer sets for PCR amplification were Reln-forward primer (50 -TGCACGACCGGTGCCATCTG-30 ) and Reln-reverse primer (50 -TTGGCTGGTGCTGGGCGATG-30 ); b-actinforward primer (50 -TCACCCACACTGTGCCCATCTACG-30 ) and b-actinreverse primer (50 -CAGCGGAACCGCTCATTGCCAA TG-30 ); mouse Reln forward primer (50 -CAAGAACAATACCGCTGATTGG-30 ) and mouse Reln reverse primer (50 -GATGTGGATGACTGTGCTCACA-30 ); mouse GAPDH forward primer (50 -AGCCTCGTCCCGTAGACAAA-30 ) and mouse GAPDH reverse primer (50 -CCTTGACTGTGCCGTTGAAT-30 ).

NaB was purchased from MERCK (Frankfurt, Germany). 3,30 ,5,50 -tetramethylbenzidine (TMB) and brillant cresyl blue were purchased from SIGMA (Saint Louis, MO). Anti-Reelin monoclonal antibody 142 was purchased from Calbiochem (San Diego, CA). The anti-mouse CD16/32, fluorescein isothiocyanate (FITC)-conjugated anti-mouse Ter119 (FITC-Ter119), FITC-conjugated antimouse IgG2b (FITC-IgG2b), and 7-aminoactinomycin D (7-AAD) were purchased from eBioscience (San Diego, CA). The R-Phycoerythrin (PE)-conjugated anti-mouse CD71 (PE-CD71) and PE-conjugated anti-mouse IgG1 (PE-IgG1) were purchased from PharMingen (San Diego, CA). The anti-b-actin antibody was from Novus Biologicals (Littleton, CO). The anti-phospho-AKT (S473) antibody was purchased from Epitomics (Burlingame, CA). The anti-AKT and anti-tubulin antibodies were from Cell Signaling (Danvers, MA). Lentivirus-based expression plasmids encoding short-hairpin interfering RNA for Reelin and luciferase were provided by the National RNAi Core Facility, Academia Sinica, Taiwan. The QuantikineÒ ELISA mouse erythropoietin immunoassay was purchased from R&D Systems (Minneapolis, MN). The Iron liquicolor and the total iron binding capacity (TIBC) kits were purchased from HUMAN Diagnostics (Wiesbaden, Germany). 2.2. Animals, genomic DNA isolation and genotyping The B6C3Fe-a/a-Relnlrl strain of heterozygous reeler mice (hereafter named Reln+/) was obtained from Jackson Laboratory (Bar Harbor, ME). Reln+/ mice were interbred to generate wild type (hereafter named Reln+/+), heterozygous and homozygous (hereafter named Reln/) reeler mice. The animal work has been reviewed and approved by the Institutional Animal Care and Use Committee. Two to three week-old mice were used in this study. Genomic DNA isolation was performed as described in the Supplementary methods, and genotyping of new pups was performed as described previously [12] and in the Supplementary methods. 2.3. Cell culture, erythroid differentiation and benzidine staining K562 cells were maintained in RPMI-1640 supplemented with 5% fetal bovine serum (FBS). Erythroid differentiation of K562 cells was induced by treatment with NaB. The extent of erythroid differentiation was determined by benzidine staining using TMB solution as described in the Supplementary methods [13]. The percentage of TMB-positive cells was determined by counting a total of 200 cells under a phase contrast microscope at 100X magnification. 2.4. Generation of shReln and shLuc stable cell lines, preparation of cell lysates and Western blot analysis Lentiviruses encoding short-hairpin interfering RNA were produced to infect K562 cells and to establish shReln or shLuc stable cell lines [14]. At 24 h post-infection, stable clones were selected for 3 weeks by addition of 1 lg/ml puromycin, and were maintained in culture medium containing 0.7 lg/ml puromycin. Pooled clones were used in this study to avoid the effects of different erythroid differentiation efficiencies arising from simple subcloning [15]. Preparation of cell lysates and Western blot analysis was performed as described previously [16]. The concentrations for the primary antibodies used in this study were anti-Reelin (1:1000), anti-phospho-AKT (S473) (1:1000), anti-AKT (1:1000), anti-b-actin (1:10000) and anti-tubulin (1:5000). The band intensity was quantified by ImageJ software (National Institute of Health).

2.6. Bone marrow sections and hematoxylin and eosin (HE) staining Mouse femurs were fixed in 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4, at 4 °C overnight. The tissues were decalcified with 15% EDTA and were embedded in paraffin. The paraffin sections (5 lm) were deparaffinised and stained with Mayer’s hematoxylin and eosin solution. 2.7. Determination of RBC-related haematological profiles Mouse whole blood was collected by retro-orbital puncture into a 0.5 ml EDTA.K3 vacuum blood collection tube. The RBC count, mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), MCHC and red blood cell distribution width (RDW) were determined using the SYSMEX XT-1800i analyzer (Kobe, Japan). 2.8. Determination of reticulocyte production index (RPI) The peripheral blood was mixed with an equal volume of 1% brilliant cresyl blue solution for reticulocyte staining. The number of reticulocytes that was revealed during the process of counting 1000 RBC was used to define the percentage of reticulocyte in the RBC population (% reticulocyte). The reticulocyte production index (RPI) was calculated according to the equation RPI = % reticulocyte  (Hct of the test sample/mean Hct of wild type), assuming that the maturation rate for the wild type and Reln/ reticulocytes is equal. 2.9. Determination of plasma EPO, plasma iron, and total iron binding capacity (TIBC) Heparinized plasma from peripheral blood collected by cardiac puncture was obtained by centrifugation at 1200g for 10 min. Plasma EPO levels were determined using the QuantikineÒ ELISA mouse erythropoietin immunoassay as described by the manufacturer (R&D Systems). Plasma iron levels were determined using the Iron liquicolor kit (HUMAN Diagnostics). For measurement of TIBC, heparinized plasma (20 ll) was added to 40 ll excess Fe(III) iron buffer and then aluminum oxide was used to absorb and precipitate unbound iron. After centrifugation, the supernatant was collected for TIBC detection (5 ll) as described by the manufacturer (HUMAN Diagnostics). 2.10. Flow cytometry analysis of cell population in the erythroid lineage The assay is based on the expression of cell-surface erythroblast-specific epitope Ter119 together with CD71 (the transferrin receptor) and the forward scatter (FSC) parameter [18]. Bone marrow cells were obtained from mouse femurs and washed with

60

H.-C. Chu et al. / FEBS Letters 588 (2014) 58–64

5% FBS–PBS. After blocking the Fc-receptor by anti-mouse CD16/ 32, the cells were incubated on ice for 1 h with FITC-Ter119 (0.25 lg) and PE-CD71 (0.2 lg) antibodies in 5% FBS–PBS. Control samples were incubated with the FITC-IgG2b and PE-IgG1 isotype control antibody. 7-AAD was used to exclude dead cells. Flow cytometry was performed using the Accuri 6 Cell Analyzer (BD Biosciences, San Diego, CA) and the distribution of cell populations was analyzed by FlowJo software (Tree star Inc., Ashland, OR). All Ter119-positive cells were classified into four subsets: ProE (Ter119medCD71high), EryA (Ter119highCD71highFSChigh), EryB (Ter119highCD71highFSClow) and EryC (Ter119highCD71lowFSClow), corresponding to the morphologically recognized proerythroblasts and basophilic, polychromatic and orthochromatic erythroblasts, respectively. Mouse progenitor cells for detection of Reln mRNA and AKT-S473 phosphorylation were sorted by FACSAria cell sorter based on the expression of Ter119 and CD71. 2.11. Statistical analysis Student’s t-test or One-way ANOVA followed by Bonferroni’s multiple comparison test was used for statistical analysis. A P < 0.05 was considered as statistically significant. 3. Results 3.1. Reelin expression is up-regulated during NaB-induced erythroid differentiation of K562 cells Because they express Reelin [5] and are able to differentiate into lineage-specific blood cells [19], K562 cells were used in this study to address the role of Reelin in erythroid differentiation. Western blot analysis revealed that the full-length 420 kDa Reelin protein was elevated by NaB in a dose-dependent manner (Fig. 1A). In addition, the 180 kDa immunoreactive band corresponding to the proteolytic product of Reelin [2] was also detectable. Time course study revealed that NaB-induced up-regulation of Reelin was sustained for more than 48 h (Fig. 1B) with 11.3 ± 2.0 (n = 3) and 12.4 ± 2.8 (n = 3) fold increase at 24 and 48 h after NaB treatment, respectively (P < 0.05). The slight increase in Reelin without sodium butyrate is likely a consequence of spontaneous erythroid differentiation [20]. These results indicate that Reelin is up-regulated during erythroid differentiation of K562 cells. To determine whether Reelin mRNA is up-regulated during NaBinduced erythroid differentiation, real-time RT-PCR was performed to quantify Reelin mRNA expression. When compared to the

untreated control cells, Reelin mRNA was increased by 2.43 ± 0.14-fold (Fig. 1C, P < 0.01, n = 3) and thus may contribute to the increase in Reelin protein. 3.2. Deficiency of Reelin enhances erythroid differentiation of K562 cells To elucidate the roles of Reelin in NaB-induced erythorid differentiation, K562 stable cell lines were established expressing Reelin (shReln) and luciferase (shLuc) short-hairpin interfering RNA. Because the presence of Reelin in FBS (Supplementary Fig. 1) can potentially mask the effect of Reelin knockdown, erythroid differentiation was performed in serum free (and thus ‘‘Reelin-free’’) OPTI-MEM to eliminate this possible interference. Under these conditions, both basal and NaB-induced Reelin expression was diminished in the shReln cells (Fig. 2A). At 48 h after differentiation induction, shReln cells had a 38% increase in TMB-positive cells when compared to shLuc cells (Fig. 2B, 26.5% for shReln vs. 19.2% for shLuc; P < 0.05, n = 10). Reelin deficiency thereby enhances erythroid differentiation of K562 cells. 3.3. Erythrocyte production is increased in homozygous reeler mice To confirm the effect of Reelin knockdown on augmenting erythroid differentiation of K562 cells, and to determine whether Reelin deficiency enhances erythroid differentiation and RBC production in vivo, RBC-related indexes for Reln+/+, Reln+/ and Reln/ mice were assessed. Reln+/+ and Reln+/ mice displayed similar RBC-related indexes (Table 1). In contrast, when compared to the Reln+/+ mice, the RBC count, Hb content, Hct and MCHC for Reln/ mice was increased by 31.3%, 25.4%, 17.2% and 7.2%, respectively (Table 1, P < 0.001). Moreover, Reln/ mice had an 11.4% decrease in MCV (P < 0.001). No difference for MCH and RDW was observed between Reln+/+ and Reln/ mice. The increase in RBC count and Hb content led us to investigate whether the polycythemia in Reln/ mice is a primary consequence of Reelin deficiency on erythroid differentiation or a secondary event driven by factors outside the erythroid compartment. The JAK2-V617F mutation, known to be associated with the acquired type of primary polycythemia [21], is not present in Reln/ mice (not shown) and in any event is on chromosome 19 and thus would not co-segregate with the reelin gene on chromosome 5. Analyses of plasma iron content (210.5 ± 34.2 mg/dl for Reln+/+ (n = 4) vs. 180.4 ± 41.8 mg/dl for Reln/ (n = 5), p = n.s.) and TIBC levels (350.9 ± 46.0 mg/dl for Reln+/+ (n = 4) vs.

Fig. 1. Reelin expression is up-regulated during NaB-induced erythroid differentiation of K562 cells. (A and B) K562 cells were treated with the indicated concentrations of NaB for 24 h (panel A) or were cultured in the presence or absence of NaB (0.5 mM) for 24 h and 48 h (panel B). Reelin expression was determined by Western blotting, with the expression of b-actin as the control for equal protein loading. The band intensity of Reelin normalized by b-actin was quantified by ImageJ software. The data represent the mean ± S.E.M. of 3 independent experiments with Reelin expression in the untreated control cells at 0 h arbitrarily set as 1 (panel B). ⁄P < 0.05. (C) K562 cells were treated with NaB or solvent control for 24 h. Relative Reelin mRNA expression in the control and NaB-treated cells was determined by real-time RT-PCR. The data represent the mean ± S.E.M. of 3 independent experiments with Reelin mRNA expression in the untreated control cell arbitrarily set as 1. C, control cells; N, NaB-treated cells. ⁄⁄P < 0.01.

H.-C. Chu et al. / FEBS Letters 588 (2014) 58–64

61

Fig. 2. Knockdown of Reelin expression enhances erythroid differentiation of K562 cells. (A and B) NaB-induced erythroid differentiation was performed in OPTI-MEM for 24 (panel B) or 48 h (panel A and B). Reelin expression was determined by Western blotting, with the expression of b-actin as the control for equal protein loading (panel A, upper panel). Data represent the mean ± S.E.M. (n = 5) for the relative levels of Reelin expression (panel A, lower panel). Reelin expression in the NaB-treated shLuc cells was arbitrarily set as 1. The extent of erythroid differentiation was determined by TMB staining (panel B, upper panel). A total of 200 cells were counted and the percentage of TMB-positive cells is shown (panel B, lower panel). The data represent the mean ± S.E.M. of 10 independent experiments. ⁄⁄P < 0.01; ⁄⁄⁄P < 0.001. Representative phase contrast microscopy images of TMB-stained cells are shown (panel B, upper panel). C, control untreated cells; N, NaB-treated cells.

Table 1 RBC profiles of Reln+/+, Reln+/ and Reln/ mice. Parametersa

Reln+/+ (n = 8)

Reln+/ (n = 10)

Reln/ (n = 6)

RBC (M/ll) Hb (g/dl) HCT (%) MCV (fl) MCH (pg) MCHC (g/dl) RDW (%)

8.0 ± 0.2 12.2 ± 0.3 46.0 ± 1.1 57.7 ± 0.4 15.3 ± 0.2 26.5 ± 0.3 20.0 ± 0.6

8.1 ± 0.1 12.2 ± 0.2 46.6 ± 0.8 57.6 ± 0.5 15.1 ± 0.2 26.2 ± 0.2 18.9 ± 1.5

10.5 ± 0.2*** 15.3 ± 0.2*** 53.9 ± 0.8*** 51.1 ± 0.4*** 14.6 ± 0.2 28.4 ± 0.4*** 22.8 ± 0.3

a One-way ANOVA followed by Bonferroni’s multiple comparison test was used for statistical analyses. Data are expressed as mean ± S.E.M. *** P < 0.001 when compared to Reln+/+ mice.

357.5 ± 34.8 mg/dl for Reln/ (n = 5), p = n.s.) revealed that these two parameters were comparable between Reln+/+ and Reln/ mice (Fig. 3A). However, Reln/ mice exhibited a significant decrease in RPI (16.4 ± 1.30 for Reln+/+ (n = 3) vs. 9.0 ± 1.1 for Reln/ (n = 3), P < 0.01) and plasma EPO levels (182.1 ± 7.0 pg/ml for Reln+/+ (n = 7) vs. 57.8 ± 5.9 pg/ml for Reln/ (n = 7), P < 0.001) when compared to the Reln+/+ mice (Fig. 3A). Hence, iron loss and augmentation of EPO production are not the driving factors leading to erythrocytosis in Reln/ mice. In contrast to the normal organ size of heart and kidney, Reln/ mice had a relatively small spleen size (0.58 ± 0.04% of body weight for Reln+/+ (n = 3) vs. 0.20 ± 0.01% for Reln/ (n = 3), P < 0.001) (Fig. 3B). Erythrocytosis in Reln/ mice is thus not likely a consequence of extramedullary erythropoiesis in spleen, which is usually associated with splenomegaly. 3.4. Reelin deficiency attenuates phosphorylation of the Reelin signaling protein AKT and alters the number and frequency of erythroid progenitors in bone marrow

Reln/ mice were slightly shorter (20%) than the femurs from Reln+/+ mice and appeared dark red (Fig. 4A). After normalization by the bone length, the number of erythrocytes, as defined by the shape of biconcave discs with no nuclei, was increased in the Reln/ mice bone marrow sections (13.0 ± 1.9 RBC/section for Reln+/+ (n = 5) vs. 175.8 ± 30.8 RBC/section for Reln/ (n = 5, P < 0.001, Fig. 4A). We therefore performed flow cytometric assays that allow direct identification and analysis of proerythroblasts and erythroblasts in freshly isolated hematopoietic tissue. The histograms illustrating the distribution of erythroid lineage bone marrow cells indicated that Reln+/+ and Reln/ mice displayed different erythropoietic activity (Fig. 4B). Reelin transcript was expressed in the Ter119+CD71+ progenitors from the Reln+/+ mice (Fig. 4C). In the absence of Reelin expression, the phosphorylation and hence the activation of the Reelin signaling protein AKT [22] was attenuated by 50% (P < 0.05, Fig. 4D). Attenuation of AKT phosphorylation associated with Reelin deficiency was further confirmed in the shReln-K562 cells that underwent erythroid differentiation (Supplementary Fig. 2). According to the histogram for the distribution of erythroid progenitors (Fig. 4B), Reln/ mice displayed a significant increase in the number of middle and late (EryB + EryC) stages of erythroid progenitors in the bone marrow (Fig. 4E, left panel). When different subsets of erythroid progenitors were analyzed, the Reln/ mice displayed a decrease in the frequency for the proerythroblasts (ProE) and the early stage (EryA) erythroblasts (P < 0.001), with a significant increase in the frequency of the middle (EryB) and late (EryC) stages of erythroblast, (P < 0.001) when compared to the Reln+/+ mice (Fig. 4E, right panel). These data imply that Reelin deficiency attenuates AKT phosphorylation and enhances RBC production in bone marrow. 4. Discussion

We determined whether the augmentation of erythroid production in Reln/ mice is related to the bone marrow erythropoiesis activity. The femurs from Reln+/+ and Reln/ mice were acquired for bone marrow section analysis and for phenotypic characterization. The gross view showed that femurs from

In the present study, we found that Reelin is up-regulated during NaB-induced erythroid differentiation of K562 cells. Rather than promoting erythorid differentiation, Reelin negatively regulates this process in erythroleukemia cells. Accordingly,

62

H.-C. Chu et al. / FEBS Letters 588 (2014) 58–64

Fig. 3. Reeler mice display an increase in RBC production independent of factors outside the erythroid compartment. (A) The plasma from Reln+/+ and Reln/ mice was collected for reticulocyte count and the analyses of EPO and iron content. Data represent the mean ± S.E.M. of 3–7 independent experiments. n.s., not significant. (B) The image (left panel) and the weight (right panel) for the indicated organs collected from the Reln+/+ and Reln/ mice are shown. Data represent the mean ± S.E.M. (n = 3) for the weight of organs after normalization by the body weight.

homozygous reeler mice exhibit a higher RBC production activity in bone marrow, with an increase in RBC count and an alteration of RBC-related indices in peripheral blood. This study thereby defines for the first time a negative regulatory role of Reelin in erythroid differentiation. Erythropoiesis is augmented by several factors including EPO, KIT ligand, oncostatin-M, and glucocorticoids [23]. However, less is understood concerning factors that limit this process. Putative negative regulators such as DYRK3 and DAPK2 have been proposed to act coordinately to control erythroid production [24]. Both DYRK3 and DAPK2 are expressed selectively in late erythroid progenitor cells in mice, and are proven to attenuate erythroblast formation and RBC production [25,26]. Experimental evidence obtained in this study demonstrates that Reelin may play a meaningful role as an erythropoietic suppressor to limit the process of erythropoiesis. The inhibitory effect of Reelin on erythroid differentiation was apparent when K562 cells were induced to differentiate in OPTI-MEM culture medium, but not in RPMI-1640 supplemented with 5% FBS. Whether the presence of growth factors or the like in FBS masks the effect of Reelin is worthy of further investigation. Despite this, the data suggest that Reelin be added to the list of regulators that negatively control erythroid differentiation. Although the upregulation of Reelin in erythroid differentiated K562 cells appears to contradict the role of Reelin as a negative regulator of erythroid differentiation, we can not rule out that differentiation signals induced by sodium butyrate may at the same time initiate a cellular process to control the extent of erythroid differentiation and hence the homeostasis of K562 cells. It is also likely that Reelin signaling and regulation is relatively sophisticated in vivo and can not be completely reflected in the K562 cell model. Similar to this scenario, the roles of ERK1/2 activation in erythroid differentiation are controversial in different in vitro and in vivo models. For example, ERK1/2 activation is indispensible for hemin- but not butyrate-induced erythroid differentiation of K562 cells [27–29]. The MAPK/ERK pathway was also found to be required in early but not in late erythroid progenitors in the selfrenewal of avian erythroid progenitors [30]. On the other hand, activation of ERK1/2 by EPO and stem cell factor is important for

expansion of human primary erythroid progenitors but is not required for the renewal of primary murine erythroid progenitors [31,32]. Moreover, ERK1/ mice display an enhanced splenic erythropoiesis without any effect on bone marrow erythropoiesis, implying that ERK1 is a negative regulator of the adult steady-state splenic erythropoiesis [33]. These studies all point out the complicated nature of the mechanisms regulating erythroid differentiation, and thus validation of Reelin function in different erythroid differentiation models is noteworthy. Consistent with the previous report [22] showing that the AKT pathway is one of the effectors in Reelin signaling, the Ter119+CD71+ progenitors from Reelin-deficient mice display a decrease in AKT-Ser473 phosphorylation. Activation of AKT through Ser473 phosphorylation is crucial for EPO-induced erythropoiesis and has fundamental roles in the regulation of cell cycle, survival and differentiation [34,35]. However, as demonstrated by the study of Breig et al., inhibition of PI3K/AKT signaling by LY294002 downregulates Spi-1/PU.1 expression and is sufficient to induce hemoglobin synthesis and commit Friend erythroleukemia cells to late erythroid differentiation [36]. Hence, AKT appears to have dual role and, depending on the stage of erythroid maturation, functions either as positive or negative regulator of erythropoiesis. Based on the observations that Reelin-deficient Ter119+CD71+ progenitors display a decrease in AKT-Ser473 phosphorylation and the enhanced erythropoietic activity in reeler mice, this study fuels the concept that, by regulating AKT phosphorylation and activation, Reelin plays a pivotal role in the final maturation of steady state erythrocytes. An increase in RBC count, Hb content, Hct, and MCHC in the homozygous reeler mice provides a further link between Reelin deficiency and polycythemia in vivo. The JAK2-V617F mutation is the common cause of human polycythemia vera, an acquired type of primary polythemia [21]. The reeler mice do not inherit this type of mutation, ruling out JAK2-V617F as the molecular basis for the polycythemia associated with Reelin deficiency. Because of the decrease in plasma EPO and the normal plasma iron and TIBC levels, the polycythemia in reeler mice is not likely driven by these factors outside the erythroid compartment. Moreover, the polycythemia does not appear as a secondary event associated with neuronal

H.-C. Chu et al. / FEBS Letters 588 (2014) 58–64

63

Fig. 4. Alteration of erythropoiesis activity in the bone marrow of Reln/ mice. (A) The femurs of Reln+/+ and Reln/ mice were collected at 3 weeks after birth for phenotypic characterization (left panel) and for HE staining of bone marrow sections (middle panel). Erythrocytes indicated by white arrows were counted and the number of RBC per bone marrow section was determined. Data (right panel) represent the mean ± S.E.M. (n = 5) for the number of RBC/section from the Reln+/+ and Reln/ mice. (B–E) Flowcytometric analysis of bone marrow cells labeled with antibodies directed against Ter119 and CD71. Ter119high cells were further analyzed with respect to FSC and CD71. The histograms illustrate the distribution of erythroid lineage bone marrow cells from Reln+/+ and Reln/ mice (panel B). The Ter119+CD71+ progenitors from Reln+/+ or Reln/ mice obtained by FACSAria sorting were analyzed for Reelin mRNA expression (panel C) and AKT phosphorylation (panel D). For AKT phosphorylation, the data represent the mean ± S.E.M. of 3 independent experiments with the phosphoryation level of AKT in Reln+/+ mice arbitrarily set to 100%. The mean ± S.E.M. for the number (n = 5, left panel) and frequency (n = 6, right panel) of the indicated erythroid progenitors are also shown (panel E). NTC, no template control. ⁄P < 0.05; ⁄⁄P < 0.01; ⁄⁄⁄P < 0.001.

defects in reeler mice. Previous study showed that MCV is increased or remains normal when secondary erythrocytosis occurs in response to various diseases and symptoms [37]. In contradiction to this scenario, reeler mice display a decrease in MCV. The changes in the frequency and distribution of proerythroblasts and erythroblasts in the bone marrow of reeler mice support the notion that Reelin deficiency affects RBC differentiation. In addition, Reelin deficiency augments NaB-induced erythroid differentiation of K562 cells, implying that Reelin indeed is involved in the regulation of erythroid differentiation. Based on the in vitro and in vivo studies, Reelin is likely to have a direct impact on erythroid differentiation. Whether polycythemia plays a role in the neurodegenerative symptom in reeler mice is unknown. Excessive erythrocytosis by overexpressing Epo in a transgenic mouse model leads to hepatic, renal, neuronal and muscular degeneration [38], implying that either the increase in RBC number or the excessive EPO (which does not occur in the reeler mice) could partially contribute to the neuronal defects of reeler mice.

Unveiling new molecules that regulate the survival and maturation of erythrocytes is clinically important and relevant to the anemia of chemotherapy, chronic myelodysplasia and multiple myeloma. The findings we report herein provide a new insight into the role of Reelin in erythroid differentiation, and contribute to our understanding of the function of extraneuronal reelin. Acknowledgements We thank Drs. Kou-Ray Lin and Jeffrey Jong-Young Yen (Academia Sinica, Taipei, Taiwan) for their technical advice on flow cytometry analysis of erythroid progenitors. This work was supported in part by the National Science Council Grant NSC 992628-B-182-001-MY3, NSC99-2632-B-182-001-MY3, 102-2628B-182-009-MY3 and 102-2628-B-182-010-MY3, Chang Gung Memorial Hospital Grants CMRPD180423 and CMRPD1B0392, and Chang Gung Molecular Medicine Research Center Grant EMRPD1C0121 to C.P.T; CMU101-S-40 to J.C.C.

64

H.-C. Chu et al. / FEBS Letters 588 (2014) 58–64

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.febslet.2013. 11.002. References [1] Maurin, J.C., Couble, M.L., Didier-Bazes, M., Brisson, C., Magloire, H. and Bleicher, F. (2004) Expression and localization of reelin in human odontoblasts. Matrix Biol. 23, 277–285. [2] Smalheiser, N.R., Costa, E., Guidotti, A., Impagnatiello, F., Auta, J., Lacor, P., Kriho, V. and Pappas, G.D. (2000) Expression of reelin in adult mammalian blood, liver, pituitary pars intermedia, and adrenal chromaffin cells. Proc. Natl. Acad. Sci. USA 97, 1281–1286. [3] Leemhuis, J., Bouché, E., Frotscher, M., Henle, F., Hein, L., Herz, J., Meyer, D.K., Pichler, M., Roth, G., Schwan, C. and Bock, H.H. (2010) Reelin signals through apolipoprotein E receptor 2 and Cdc42 to increase growth cone motility and filopodia formation. J. Neurosci. 30, 14759–14772. [4] Kim, H.M., Qu, T., Kriho, V., Lacor, P., Smalheiser, N., Pappas, G.D., Guidotti, A., Costa, E. and Sugaya, K. (2002) Reelin function in neural stem cell biology. Proc. Natl. Acad. Sci. USA 99, 4020–4025. [5] Tseng, W.L., Huang, C.L., Chong, K.Y., Liao, C.H., Stern, A., Cheng, J.C. and Tseng, C.P. (2010) Reelin is a platelet protein and functions as a positive regulator of platelet spreading on fibrinogen. Cell. Mol. Life Sci. 67, 641–653. [6] Jacquel, A., Herrant, M., Legros, L., Belhacene, N., Luciano, F., Pages, G., Hofman, P. and Auberger, P. (2003) Imatinib induces mitochondria-dependent apoptosis of the Bcr-Abl-positive K562 cell line and its differentiation toward the erythroid lineage. FASEB J. 17, 2160–2162. [7] Ikeda, Y. and Terashima, T. (1997) Expression of reelin, the gene responsible for the reeler mutation, in embryonic development and adulthood in the mouse. Dev. Dyn. 210, 157–172. [8] Redmond, L.C., Dumur, C.I., Archer, K.J., Haar, J.L. and Lloyd, J.A. (2008) Identification of erythroid-enriched gene expression in the mouse embryonic yolk sac using microdissected cells. Dev. Dyn. 237, 436–446. [9] Redmond, L.C., Dumur, C.I., Archer, K.J., Grayson, D.R., Haar, J.L. and Lloyd, J.A. (2011) Kruppel-like factor 2 regulated gene expression in mouse embryonic yolk sac erythroid cells. Blood Cells Mol. Dis. 47, 1–11. [10] Andersson, L.C., Jokinen, M. and Gahmberg, C.G. (1979) Induction of erythroid differentiation in the human leukaemia cell line K562. Nature 278, 364–365. [11] Hamburgh, M. (1963) Analysis of the postnatal developmental effects of ‘‘Reeler’’, a neurological mutation in mice. A study in developmental genetics. Dev. Biol. 19, 165–185. [12] D’Arcangelo, G., Miao, G.G. and Curran, T. (1996) Detection of the reelin breakpoint in reeler mice. Brain Res. Mol. Brain Res. 39, 234–236. [13] Bergh, G., Ehinger, M., Olofsson, T., Baldetorp, B., Johnsson, E., Brycke, H., Lindgren, G., Olsson, I. and Gullberg, U. (1997) Altered expression of the retinoblastoma tumor-suppressor gene in leukemic cell lines inhibits induction of differentiation but not G1-accumulation. Blood 89, 2938–2950. [14] Hung, W.S., Huang, C.L., Fan, J.T., Huang, D.Y., Yeh, C.F., Cheng, J.C. and Tseng, C.P. (2012) The endocytic adaptor protein disabled-2 is required for cellular uptake of fibrinogen. Biochim. Biophys. Acta 1823, 1778–1788. [15] Axelrod, D.E., Terry, R. and Kern, F.G. (1979) Cell differentiation rates of Friend murine erythroleukemia variants isolated by sib selection. Somatic Cell Genet. 5, 539–549. [16] Tseng, C.P., Huang, C.H., Tseng, C.C., Lin, M.H., Hsieh, J.T. and Tseng, C.H. (2001) Induction of disabled-2 gene during megakaryocyte differentiation of K562 cells. Biochem. Biophys. Res. Commun. 285, 129–135. [17] Livak, K.J. and Schmittgen, T.D. (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2DDCt method. Methods 25, 402–408. [18] Socolovsky, M. (2007) Molecular insights into stress erythropoiesis. Curr. Opin. Hematol. 14, 215–224. [19] Sutherland, J.A., Turner, A.R., Mannoni, P., McGann, L.E. and Turc, J.M. (1986) Differentiation of K562 leukemia cells along erythroid, macrophage, and megakaryocyte lineages. J. Biol. Response Mod. 5, 250–262.

[20] Koeffler, H.P. (1986) Human acute myeloid leukemia lines: models of leukemogenesis. Semin. Hematol. 23, 223–236. [21] Tefferi, A. (2011) Annual clinical updates in hematological malignancies: a continuing medical education series: polycythemia vera and essential thrombocythemia: 2011 update on diagnosis, risk-stratification, and management. Am. J. Hematol. 86, 292–301. [22] Jossin, Y. and Goffinet, A.M. (2007) Reelin signals through phosphatidylinositol 3-kinase and Akt to control cortical development and through mTor to regulate dendritic growth. Mol. Cell. Biol. 27, 7113–7124. [23] Bogacheva, O., Bogachev, O., Menon, M., Dev, A., Houde, E., Valoret, E.I., Prosser, H.M., Creasy, C.L., Pickering, S.J., Grau, E., Rance, K., Livi, G.P., Karur, V., Erickson-Miller, C.L. and Wojchowski, D.M. (2008) DYRK3 dual-specificity kinase attenuates erythropoiesis during anemia. J. Biol. Chem. 283, 36665– 36675. [24] Wojchowski, D.M., Menon, M.P., Sathyanarayana, P., Fang, J., Karur, V., Houde, E., Kapelle, W. and Bogachev, O. (2006) Erythropoietin-dependent erythropoiesis: new insights and questions. Blood Cells Mol. Dis. 36, 232– 238. [25] Geiger, J.N., Knudsen, G.T., Panek, L., Pandit, A.K., Yoder, M.D., Lord, K.A., Creasy, C.L., Burns, B.M., Gaines, P., Dillon, S.B. and Wojchowski, D.M. (2001) MDYRK3 kinase is expressed selectively in late erythroid progenitor cells and attenuates colony-forming unit-erythroid development. Blood 97, 901– 910. [26] Fang, J., Menon, M., Zhang, D., Torbett, B., Oxburgh, L., Tschan, M., Houde, E. and Wojchowski, D.M. (2008) Attenuation of EPO-dependent erythroblast formation by death-associated protein kinase-2. Blood 112, 886–890. [27] Woessmann, W. and Mivechi, N.F. (2001) Role of ERK activation in growth and erythroid differentiation of K562 cells. Exp. Cell Res. 264, 193–200. [28] Woessmann, W., Zwanzger, D. and Borkhardt, A. (2004) ERK signaling pathway is differentially involved in erythroid differentiation of K562 cells depending on time and the inducing agent. Cell Biol. Int. 28, 403–410. [29] Witt, O., Sand, K. and Pekrun, A. (2000) Butyrate-induced erythroid differentiation of human K562 leukemia cells involves inhibition of ERK and activation of p38 MAP kinase pathways. Blood 95, 2391–2396. [30] Dazy, S., Damiola, F., Parisey, N., Beug, H. and Gandrillon, O. (2003) The MEK-1/ ERKs signalling pathway is differentially involved in the self-renewal of early and late avian erythroid progenitor cells. Oncogene 22, 9205–9216. [31] Sui, X., Krantz, S.B., You, M. and Zhao, Z. (1998) Synergistic activation of MAP kinase (ERK1/2) by erythropoietin and stem cell factor is essential for expanded erythropoiesis. Blood 92, 1142–1149. [32] von Lindern, M., Deiner, E.M., Dolznig, H., Parren-Van Amelsvoort, M., Hayman, M.J., Mullner, E.W. and Beug, H. (2001) Leukemic transformation of normal murine erythroid progenitors: v- and c-ErbB act through signaling pathways activated by the EpoR and c-Kit in stress erythropoiesis. Oncogene 20, 3651–3664. [33] Guihard, S., Clay, D., Cocault, L., Saulnier, N., Opolon, P., Souyri, M., Pagès, G., Pouysségur, J., Porteu, F. and Gaudry, M. (2010) The MAPK ERK1 is a negative regulator of the adult steady-state splenic erythropoiesis. Blood 115, 3686– 3694. [34] Myklebust, J.H., Blomhoff, H.K., Rusten, L.S., Stokke, T. and Smeland, E.B. (2002) Activation of phosphatidylinositol 3-kinase is important for erythropoietininduced erythropoiesis from CD34(+) hematopoietic progenitor cells. Exp. Hematol. 30, 990–1000. [35] Bouscary, D., Pene, F., Claessens, Y.E., Muller, O., Chretien, S., Fontenay-Roupie, M., Gisselbrecht, S., Mayeux, P. and Lacombe, C. (2003) Critical role for PI 3kinase in the control of erythropoietin-induced erythroid progenitor proliferation. Blood 101, 3436–3443. [36] Breig, O., Théoleyre, O., Douablin, A. and Baklouti, F. (2010) Subtle distinct regulations of late erythroid molecular events by PI3K/AKT-mediated activation of Spi-1/PU.1 oncogene autoregulation loop. Oncogene 29, 2807– 2816. [37] Lawrence, J.H. and Berlin, N.I. (1952) Relative polycythemia; the polycythemia of stress. Yale J. Biol. Med. 24, 498–505. [38] Heinicke, K., Baum, O., Ogunshola, O.O., Vogel, J., Stallmach, T., Wolfer, D.P., Keller, S., Weber, K., Wagner, P.D., Gassmann, M. and Djonov, V. (2006) Excessive erythrocytosis in adult mice overexpressing erythropoietin leads to hepatic, renal, neuronal, and muscular degeneration. Am. J. Physiol. Regul. Integr. Comp. Physiol. 291, R947–R956.